DORA stands for DevOps Research & Assessment.
It is a research program that seeks to understand the capabilities that drive software delivery and operations performance.
It started in 2013. A few years later, Google Cloud acquired it. Every year, they publish a report based on conducting extensive surveys across the software industry. Their two most recent reports (2024 and 2025) are focused on AI.
DORA has repeatedly validated four software delivery metrics that provide an effective way of measuring the performance of the teams:
1- Change lead time: the time it takes for a code commit or change to be successfully deployed to production.
2- Deployment frequency: how often application changes are deployed to production.
3- Change fail rate: the percentage of deployments that cause failures in production, requiring hotfixes or rollbacks.
4- Failed deployment recovery time: the time it takes to recover from a failed deployment.
They also have a table for the performance level based on these 4 metrics. Teams can be in four levels of performance: Elite, High, Medium, and Low.
In this blog post, I want to show you how to leverage AI to define improvements of your product delivery process based on the DORA Metrics data.
When you want to use AI, you need to effectively communicate your request with the AI system. A good communication guideline has the following steps:
1- Teach the AI system about DORA metrics:
Teach the AI system what it is and introduce 4 metrics with their descriptions.
2- Tell it the current value of the metrics:
Tell the AI system what you have achieved regarding the 4 metrics.
3- Tell it your desired performance and value for the 4 metrics
Tell the AI system, for example, you want to become an Elite Scrum Team + the desired value for the 4 metrics. Then ask it to generate improvements to move toward that desired state.
The result of this process is the following prompt:
--------------------------------------------------
DORA metrics are 4 metrics as follows:
1- Change lead time: the time it takes for a code commit or change to be successfully deployed to production.
2- Deployment frequency: how often application changes are deployed to production.
3- Change fail rate: the percentage of deployments that cause failures in production, requiring hotfixes or rollbacks.
4- Failed deployment recovery time: the time it takes to recover from a failed deployment.
In our current product delivery process, we have achieved the following average results regarding DORA metrics as follows:
Change lead time: [Current Change lead time]
Deployment frequency: [Current Deployment frequency]
Change fail rate: [Current Change fail rate]
Failed deployment recovery time: [Current Failed deployment recovery time]
Now, we want to improve our process to become an Elite Scrum Team with the following expectations:
Change lead time: [Desired Change lead time]
Deployment frequency: [Desired Deployment frequency]
Change fail rate: [Desired Change fail rate]
Failed deployment recovery time: [Desired Failed deployment recovery time]
Create a list of improvements to help us achieve the desired results of the four DORA metrics.
--------------------------------------------------
Copy this prompt, paste it into an AI system like ChatGPT, fill in the brackets with the specific data of your context, and generate a list of improvements. It gives you great inputs for the Retrospective conversations.
This is a sample prompt:
DORA metrics are 4 metrics as follows:
1- Change lead time: the time it takes for a code commit or change to be successfully deployed to production.
2- Deployment frequency: how often application changes are deployed to production.
3- Change fail rate: the percentage of deployments that cause failures in production, requiring hotfixes or rollbacks.
4- Failed deployment recovery time: the time it takes to recover from a failed deployment.
In our current product delivery process, we have achieved the following average results regarding DORA metrics as follows:
Change lead time: Around 3 days
Deployment frequency: Once per week
Change fail rate: 15%
Failed deployment recovery time: Less than one day
Now, we want to improve our process to become an Elite Scrum Team with the following expectations:
Change lead time: Less than one day
Deployment frequency: On demand (multiple deploys per day)
Change fail rate: 5%
Failed deployment recovery time: Less than one hour
Create a list of improvements to help us achieve the desired results of the four DORA metrics.
--------------------------------------------------
Copy this prompt, paste it into an AI system like ChatGPT, and enjoy the result.
If you are a Product Owner and want to learn how to effectively leverage AI for the new generation of product management, you can attend my upcoming Professional Scrum Product Owner – AI Essentials class. Click here to see the class information.